In a study examining differential gene expression in bone from normal and osteopetrotic (op) rats, we discovered a novel cDNA, termed osteoactivin (OA) that was over-expressed in op when compared to normal bone. Subsequent in situ hybridization and immunohistochemistry studies demonstrated that OA mRNA and protein are expressed by osteoblasts. In primary osteoblast cultures, OA mRNA levels exhibited a temporal pattern of expression being expressed at highest levels during the later stages of matrix maturation and mineralization. Furthermore, the protein is synthesized by osteoblasts and secreted into the medium. Using an OA-anti-sense oligonucleotide, we were able to block OA expression. Down-regulation of OA expression inhibited osteoblast differentiation and function including decreased alkaline phosphatase activity, osteocalcin production and matrix mineralization. A CMV-rOA construct was generated and used to examine the effect of OA over-expression on osteoblast development and function in osteoblast cultures. OA over-expression increased nodule formation, alkaline phosphatase activity, osteocalcin production and matrix mineralization. Using a local injection model to test the anabolic effect of a small peptide of OA (OA-P) in vivo, OA-P induced a potent osteogenic response. We hypothesize that OA is an anabolic bone growth factor. We propose that OA is synthesized and secreted by osteoblasts, where it acts as an ECM-associated signaling molecule or a soluble factor to promote osteoblast differentiation and function Experiments proposed in this application will focus on the association between OA and bone with the following aims. Experiments in aim I will provide a more comprehensive evaluation on the effects of down-regulation of OA expression using an OA-anti-sense oligo transfection approach. OA over-expression will be achieved using a CMV-rOA construct and its effects on osteoblast development and function will be evaluated. Studies in aim 2 will examine the effects of recombinant OA (rOA) or OA-P on osteoblast development and function in vitro and bone formation in vivo. Studies in aim 3 will characterize rOA binding to osteoblasts and determine whether integrins serve as receptors for OA on osteoblasts. In aim 4, studies will examine transcriptional regulation of OA by three osteotropic factors, 1,25(OH)2D3, BMP-2 and TGF-beta1. Additional experiments will determine whether specific effects of BMP-2 on osteoblast differentiation are OA-dependent. Experiments in the first part of aim 5 will focus on the generation of OA null (-/-) mice and characterization of their skeletal phenotype. We will also examine the ability of OA -/- osteoblasts to differentiate in vitro. In the final part of this aim, we propose to generate transgenic mice expressing OA under control of the osteocalcin promoter and evaluate whether osteoblast-specific expression has an anabolic effect on bone. The identification of a novel anabolic agent in bone and elucidation of its mechanism(s) of action will eventually lead to the development of new therapeutic strategies to selectively stimulate osteogenesis in diseases associated with bone loss and in fracture repair.